Analytical Biochemistry 451 (2014) 63–68

Contents lists available at ScienceDirect

Analytical Biochemistry journal homepage: www.elsevier.com/locate/yabio

A human kringle domain-based fluorescence-linked immunosorbent assay system Gu Min Jeong a, Yong Sung Kim b, Ki Jun Jeong a,c,⇑ a

Department of Chemical and Biomolecular Engineering, Korea Advanced Institute of Science and Technology (KAIST), Yuseong-gu, Daejeon 305-701, Republic of Korea Department of Molecular Science and Technology, Ajou University, Yeongtong-gu, Suwon 443-749, Republic of Korea c Institute for the BioCentury, Korea Advanced Institute of Science and Technology (KAIST), Yuseong-gu, Daejeon 305-701, Republic of Korea b

a r t i c l e

i n f o

Article history: Received 6 November 2013 Received in revised form 4 January 2014 Accepted 31 January 2014 Available online 10 February 2014 Keywords: Kringle domain FLISA GFP Immunoassay

a b s t r a c t As a non-immunoglobulin protein scaffold, human kringle domain (KD) has attractive properties such as high specificity, stability, and production in bacterial hosts. Here, we developed a rapid and sensitive fluorescence-linked immunosorbent assay (FLISA) system using a fluorescent kringle domain (fluoKD), a fusion protein of a green fluorescent protein (GFP), and a kringle domain variant (KD548). Two kinds of fluoKDs in which KD was fused to the N terminus of GFP (N-fluoKD) or the C terminus of GFP (C-fluoKD) were constructed and characterized. In Escherichia coli host, both fluoKDs were produced in high yield and solubility and were successfully purified by a simple procedure. The purified fluoKDs exhibited strong fluorescent activities and high affinities to the target antigen. Furthermore, it was successfully demonstrated that the FLISA with purified fluoKDs allowed for more rapid detection of target antigens with higher sensitivity compared with conventional enzyme-linked immunosorbent assay (ELISA), indicating that a simple, rapid, and sensitive immunoassay system could be developed by using KD instead of antibody or antibody fragments. Ó 2014 Elsevier Inc. All rights reserved.

Currently, enzyme-linked immunosorbent assays (ELISAs)1 are widely used for the detection of specific molecules, including proteins, peptides, and sugars, as well as various other small molecules and chemicals. In addition, the immunosorbent assay is an important tool in most protein-based diagnostic platforms [1]. In most immunosorbent assay systems, antibodies are used as capturing reagents because of their high specificity and affinity to target molecules. However, the intrinsic properties of antibodies also create several limitations in the immunosorbent assay systems. In many cases, antibodies are produced in a mammalian host and the production cost is high due to their low productivity and use of expensive ⇑ Corresponding author at: Department of Chemical and Biomolecular Engineering, Korea Advanced Institute of Science and Technology (KAIST), Yuseong-gu, Daejeon 305-701, Republic of Korea. Fax: +82 42 350 3910. E-mail address: [email protected] (K.J. Jeong). 1 Abbreviations used: ELISA, enzyme-linked immunosorbent assay; KD, kringle domain; FLISA, fluorescence-linked immunosorbent assay; DR5, death receptor 5; fluoKD, fluorescent kringle domain; GFP, green fluorescent protein; N-fluoKD, fluoKD in which anti-DR5 KD548 was fused to the N terminus of GFP; C-fluoKD, fluoKD in which anti-DR5 KD548 was fused to the C terminus of GFP; PCR, polymerase chain reaction; IPTG, isopropyl-b-D-thiogalactopyranoside; PBS, phosphate-buffered saline; PBST, PBS containing 0.05% Tween 20; PBSM, PBS containing 10 wt% skim milk; HRP, horseradish peroxidase; TMB, tetramethylbenzidine; SDS–PAGE, sodium dodecyl sulfate–polyacrylamide gel electrophoresis; LOD, limit of detection. http://dx.doi.org/10.1016/j.ab.2014.01.019 0003-2697/Ó 2014 Elsevier Inc. All rights reserved.

media. Even though bacterial hosts can be used for production of antibody proteins, the folding efficiency and purification yield are often poor. In addition, the instability of antibodies in harsh conditions makes them difficult to use in diagnostic systems [2]. To overcome the limitations of antibodies, exploration of alternative protein reagents with the ability of specific binding to target molecules has been stimulated and has led to the development of non-immunoglobulin protein scaffolds. Compared with typical antibodies, non-immunoglobulin protein scaffolds, used as an alternative to antibody proteins, have several advantages: (i) smaller size (5 g/L in E. coli host); therefore, the low production cost may be more beneficial when compared with most antibody molecules. In most diagnostic systems based on immunosorbent reaction, the sensitivity and stability are highly dependent on the capturing molecules immobilized on the surface of the diagnostic system. As described earlier, KD has ideal properties for diagnostic systems, and its application toward the development of diagnostic systems is highly desired. In this study, we developed a fluorescence-linked immunosorbent assay (FLISA) system using a KD variant (Fig. 1B). As a model KD variant, KD548 was used and engineered to have high affinity to human death receptor 5 (DR5), a receptor protein present on the surface of cancer cells [6]. For the fluorescent signal, two fluorescent kringle domains (fluoKDs) were constructed where anti-DR5 KD548 was fused to the N terminus or C terminus of green fluorescent protein (GFP) (N-fluoKD or C-fluoKD, respectively) with a flexible (Gly4Ser)2 linker between GFP and KD548.

After production and purification in E. coli cultivation, both fluoKDs were applied to the FLISA system and their fluorescent intensity and binding activity were analyzed. We also successfully demonstrated that the fluoKD-based FLISA system has high sensitivity compared with a conventional ELISA system. Materials and methods Bacterial strains and plasmid All bacterial strains used in this work are listed in Table 1. E. coli XL1-Blue was used as a host cell for gene manipulation and plasmid maintenance. E. coli SHuffle Express, which is engineered to promote disulfide bond formation in the cytoplasm by constitutive expression of DsbC in cytoplasm, was used for production of both fluoKDs and GFP. Even though KD has three disulfide bonds, the correctly folded KD can be produced in cytoplasm of this strain. Polymerase chain reaction (PCR) was performed with the C1000 Thermal Cycler (Bio-Rad, Hercules, CA, USA) using PrimeSTAR HS Polymerase (Takara Bio, Shiga, Japan). The nucleotide sequences of all primers used in this study are listed in Table S1 of the online Supplementary material. A GFP gene was amplified from pGFPmut2 [9] by PCR with primers CF-F4, CF-F5, CF-F6, and GFP-R1. The PCR product was digested with restriction enzymes XbaI and HindIII and then cloned into the same enzyme sites of pMoPac1 [10], yielding pGM-GFP. For the expression of N-fluoKD in which

B

A

N-fluoKD

GFP

N

Flexible linker N C

C

Target

C-fluoKD

N-fluoKD

Fig.1. (A) Schematic structure of KD (PDB entry code: 1I5K). It has a rigid core via three disufide bonds and seven flexible loops. (B) Concept of KD-based FLISA system. The fluoKDs in which KD is linked to the N or C terminus of GFP can bind to target antigens with specific affinity, and the fused GFP gives fluorescent signal for detection of target molecules.

Table 1 Bacterial strains and plasmids used in this study. E. coli strains or plasmids E. coli strains XL1-Blue BL21(DE3) SHuffle Express Plasmids pMoPacI pET21b-DR5 pPICZa-KD548 pGFPmut2 pGM-GFP pGM-N-fKD pGM-C-fKD a b c

Description

References

(recA1 endA1 gyrA96 thi-1 hsdR17 supE44 relA1 lac [F0 proAB lacIqZDM15 Tn10 (Tetr)] F– ompT gal dcm lon hsdSB(r-B m-B) k(DE3 [lacI lacUV5-T7 gene 1 ind1 sam7 nin5]) fhuA2 [lon] ompT ahpC gal katt::pNEB3-r1-cDsbC (SpecR, lacIq) DtrxB sulA11 R(mcr-73::miniTn10–TetS)2 [dcm] R(zgb-210::Tn10-TetS) endA1 Dgor D(mcrC-mrr)114::IS10

Stratagenea Novageneb New England Biolabsc

ColE1 origin, Cmr, lac promoter, lacIq 6His tag-DR5 human KD548 gene GFPmut2 gene pMoPac1 derivative, 6His tag-GFP pMoPac1 derivative, 6His tag-N-fluoKD pMoPac1 derivative, 6His tag-C-fluoKD

[10] [4] [4] [9] This study This study This study

Stratagene, Santa Clara, CA, USA. Novagene, Madison, WI, USA. New England Biolabs, Ipswich, MA, USA.

Human kringle domain-based FLISA system / G.M. Jeong et al. / Anal. Biochem. 451 (2014) 63–68

KD548 was linked to the N terminus of GFP, KD548 coding gene was amplified from pPICZa-KD548 [4] by PCR with primers NFF1, NF-F2, and NF-R2, and GFP coding gene was amplified from pGFPmut2 by PCR with primers NF-F3 and NF-R3. Then, each PCR product was mixed and N-fluoKD gene was synthesized by overlap PCR with primers NF-F2 and NF-R3. The resulting PCR product in which KD548 and GFP genes were linked via (Gly4Ser)2 linker was digested with restriction enzymes XbaI and HindIII and then cloned into pMoPac1, yielding pGM-N-fKD. For the expression of C-fluoKD in which KD548 was linked to the C terminus of GFP, KD548 coding gene was amplified by PCR with primers CF-F7 and CF-R5, and GFP coding gene was amplified by PCR with primers CF-F4, CF-F5, CF-F6, and CF-R4. Then, each PCR product was mixed and C-fluoKD gene was synthesized by overlap PCR with primers CF-F6 and CF-R5. The resulting PCR was digested with XbaI and HindIII and then cloned into pMoPac1, yielding pGM-C-fKD. The schematic diagram of plasmids used for this work is shown in Fig. S1 of the Supplementary material. All DNA manipulations, including restriction digestion, ligation, and agarose gel electrophoresis, were carried out using standard procedures [11].

65

FACS analysis of GFP-producing cell The cultured cells were harvested by centrifugation at 4 °C and 6000 rpm for 10 min. The cells were washed with 1 PBS and applied to FACS analysis. MoFlo XDP (Beckman Coulter, Brea, CA, USA) was used for FACS analysis. The fluorescence reading was obtained using an excitation of 488 nm with an argon laser. The emission signal was measured in FL1 channel centered at 530/40 nm on excitation. The sample mean fluorescence intensity values and images were analyzed using SUMMIT software version 5.2. Coulter Isoton II Diluent (Beckman Coulter) was used in all experiments as the flow cytometry sheath fluid. Measurement of fluorescence intensity of fluoKD After purification of fluoKDs and GFP, their fluorescence intensities were measured. The same concentration (4 lM, 50 ll/well) of purified GFP and fluoKDs was loaded on the 96-well black plate (Corning, Tewksbury, MA, USA). The fluorescence intensities of samples were measured by a microplate reader (Infinite M200 PRO, TECAN, Grodig, Austria). The excitation and emission wavelengths used for detection were 488 and 535 nm, respectively.

Flask cultivation Conventional ELISA For the production of both fluoKDs and GFP, E. coli SHuffle Express harboring each expression plasmid was inoculated and cultivated at 37 °C in liquid LB medium (BD, Franklin Lakes, NJ, USA) with 2% glucose. After an overnight cultivation, 5 ml of the inoculums was transferred into 500 ml of fresh medium in a 2-L flask and incubated at 37 °C while shaking at 200 rpm. When the cell density (OD600nm) reached approximately 0.6, cells were induced with isopropyl-b-D-thiogalactopyranoside (IPTG; Sigma–Aldrich, St. Louis, MO, USA) to a final concentration of 1 mM. After induction, cells were further cultivated at 37 °C for 4 h. For the production of DR5, E. coli BL21(DE3) harboring pET21bDR5 [4] was used. Cells were inoculated into the LB medium with 2% glucose and overnight cultured at 37 °C and 200 rpm. The next morning, the cells were transferred to fresh LB medium and cultivated at 37 °C. When the cell density (OD600nm) reached approximately 0.6, cells were induced with IPTG to a final concentration of 0.5 mM. The cells were further incubated at 25 °C and 200 rpm for 16 h, and then cells were harvested by centrifugation (6000 rpm, 4 °C, 10 min).

Protein purification Following flask cultivation, cells were harvested by centrifugation at 6000 rpm and 4 °C for 10 min. Then, cell pellets were resuspended in 40 ml of the binding buffer (50 mM phosphate–KOH and 300 mM NaCl, pH 8.0) and disrupted by three cycles of sonication (each for 10 min of on-time at 20% of maximum output; High-Intensity Ultrasonic Liquid Processors, Sonics & Material, Newtown, CT, USA). After cell disruption, only the soluble lysate was collected by centrifugation (10,000 rpm, 4 °C, 10 min) and was incubated with 1 ml of Ni-NTA (nickel–nitrilotriacetic acid) resin (GE Healthcare, Buckinghamshire, UK) for 1 h with shaking at 4 °C. After washing the resin with washing buffer (50 mM phosphate–KOH, 300 mM NaCl, and 5 mM imidazole, pH 8.0), proteins were eluted with elution buffer (50 mM phosphate–KOH, 300 mM NaCl, and 250 mM imidazole, pH 8.0). After dialysis against phosphate-buffered saline (PBS: 135 mM NaCl, 2.7 mM KCl, 4.3 mM Na2HPO4, and 1.4 mM KH2PO4, pH 7.2), the purified proteins were stored at 4 °C for future use. The purified proteins were quantified using Bradford assay solution (Bio-Rad). All protein purifications were carried out at 4 °C.

First, 2 lM, 50 ll/well KD or fluoKDs were coated on the 96well plate at 37 °C for 1 h and washed with 200 ll/well PBST (PBS containing 0.05% Tween 20) four times. In addition, those wells were blocked by 200 ll/well PBSM (PBS containing 10 wt% skim milk) at 37 °C for 1 h and washed with PBST four times. Next, serially diluted DR5 was applied with 50 ll/well volume to the wells and then incubated at 37 °C for 1 h. The wells were washed with PBST four times, and then 100 ll/well of 1:2000 diluted horseradish peroxidase (HRP)-conjugated anti-FLAG tag–antibody was added. After 1 h of incubation, the wells were washed with PBST four times and 50 ll/well of the TMB (tetramethylbenzidine) peroxidase substrate (BD Biosciences, San Jose, CA, USA) was added for peroxidase reaction. Finally, 50 ll of 2 M H2SO4 was added into each well to stop the peroxidase reaction and, the absorbance at 450 nm was measured by a microplate reader (Infinite M200 PRO). Conventional FLISA First, 3 lg of DR5 was coated on each well at 37 °C for 1 h, and then wells were washed with 200 ll/well PBST four times. The wells were blocked by 200 ll/well PBSM at 37 °C for 1 h and then washed with PBST four times. Next, serially diluted fluoKDs were applied with 50 ll/well volume to the wells and then incubated at 37 °C for 1 h. The wells were washed with PBST four times, and finally 50 ll/well PBS was applied. The fluorescent signal was measured by a microplate reader (Infinite M200 PRO). The excitation and emission wavelengths were 488 and 535 nm, respectively. Competitive FLISA First, wells were coated with 3 lg/well DR5 at 37 °C for 1 h and then washed with 200 ll/well PBST four times. The wells were blocked by 200 ll/well PBSM at 37 °C for 1 h and then washed with PBST four times. During these steps, 25 ll of serial diluted DR5 (from 0.64 lg/ml) in PBS was incubated with 25 ll of each fluoKD (each 400 lg/ml) solution for 1 h at 37 °C. The mixed samples were loaded on the DR5-coated wells. After incubation at 37 °C for 1 h, each well was washed with 200 ll/well PBST four times to remove the unbound fluoKDs. Finally, 50 ll/well PBS was added into the wells, and then the fluorescence signal was measured by a

Human kringle domain-based FLISA system / G.M. Jeong et al. / Anal. Biochem. 451 (2014) 63–68

microplate reader (Infinite M200 PRO). The excitation and emission wavelengths were 488 and 535 nm, respectively. Competitive ELISA Competitive ELISA was performed in microtiter well plates (Nunc, Invitrogen, Carlsbad, CA, USA). First, the wells were coated with 3 lg/well DR5 at 37 °C for 1 h and washed with 200 ll/well PBST four times. The wells were blocked by 200 ll/well PBSM at 37 °C for 1 h and washed with PBST four times. During the steps above, 25 ll of serial diluted DR5 (from 400 lg/ml) in PBS was incubated with 25 ll of each fluoKD (each 400 lg/ml) solution for 1 h at 37 °C. The mixed samples were reacted with the wells at 37 °C for 1 h, and the wells were washed with 200 ll/well PBST four times. The washed wells were blocked again with PBSM at 37 °C for 1 h, and then the plates were reprobed by 100 ll/well anti-GFP–HRP-conjugated antibody (ABCam, Cambridge, UK) in PBST (1:2000) for 1 h at 37 °C. The plates were washed four times by PBST, and 50 ll of the TMB peroxidase substrate was added for peroxidase reaction. Finally, 50 ll of 2 M H2SO4 was used to stop the peroxidase reaction. The absorbance at 450 nm was estimated using an Infinite M200 PRO microplate reader. Results Production and purification of fluoKDs

1

2

3

After purification, protein concentrations were normalized and then the fluorescence intensity of each purified fluoKD was analyzed and compared with that of wild-type GFP. Compared with wild-type GFP, N-fluoKD and C-fluoKD exhibited relatively lower intensities (82 and 72% of GFP intensity, respectively; see Fig. 3A), but both were high enough to be used for fluorometric detection. Between both fluoKDs, N-fluoKD showed slightly higher intensity than C-fluoKD (1.1-fold). This result indicates that fusion of KD to GFP at the N- and C-terminal ends did not have any deleterious effect on the fluorescent activity of GFP. The binding activities of both fluoKDs were also analyzed by ELISA and compared with that of original KD (no fusion with GFP). We clearly observed that all examined proteins except BSA (negative control) showed very similar binding activities against DR5 (Fig. 3B). As described in the introductory paragraphs, KD has a rigid core structure and so the binding activity of KD to target molecules is not easily modified by the fusion with other proteins.

4

5 6

200 150 100 75 50 37

25

To confirm the activities of N- and C-fluoKDs, the conventional FLISA was carried out. A 96-well black microtiter plate was coated with DR5 (antigen of KD548), and serially diluted fluoKDs (from 0.0128 to 40 ng) were loaded into each well and their fluorescent intensities were analyzed. In the case of N-fluoKD, the fluorescent

A Fluorescence intensity (Arbitrary unit)

M

Fluorescent intensity of fluoKD

Conventional FLISA with fluoKD

For the production of both fluoKDs (anti-DR5 KD548 fused GFPs), the constructed plasmids, pGM-N-fKD and pGM-C-fKD, were transformed into E. coli SHuffle Express for production. After flask cultivation, the production yields and solubility of each fluoKD were analyzed by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE). In the cytoplasm of E. coli, both fluoKDs were highly expressed (N-fluoKD, 17% of total proteins; C-fluoKD, 8% of total proteins) with high solubility (>90%) (Fig. 2). The production and fluorescent signal intensities of both fluoKDs in E. coli cells were also analyzed by a flow cytometer. Similar to the above expression data (Fig. 2), N-fluoKD also exhibited much higher fluorescent signal intensity than C-fluoKD (see Fig. S2 in Supplementary material). With cultured cells, purification of both fluoKDs was carried out as described in Materials and Methods, and we observed that both fluoKDs with a molecular mass of 38 kDa were successfully purified (Fig. 2). From a 1-L culture,

(kDa)

approximately 20.5 mg of N-fluoKD and 7.0 mg of C-fluoKD could be purified with high purity (>90%) and high recovery yields (40– 50%). Between both fluoKDs, N-fluoKD showed a relatively higher expression level and final production yield compared with C-fluoKD (Fig. 2).

12000 10000 8000 6000 4000 2000 0 GFP

N-fluoKD

C-fluoKD

PBS

B Absorbance at 450 nm

66

20 Concentration of DR5 (µ µg/ml) Fig.2. SDS–PAGE analysis of production and purification of N-fluoKD (lanes 1–3) and C-fluoKD (lanes 4–6). Lane M: molecular weight markers (kDa); lanes 1 and 4: total fraction; lanes 2 and 5: soluble fraction; lanes 3 and 6: purified protein. Open and closed arrowheads indicate N-fluoKD and C-fluoKD, respectively.

Fig.3. Characterization of fluoKDs. (A) Measurement of fluorescent intensities of fluoKDs. The same amount of GFP and fluoKDs was applied to measurement, and PBS without protein was used as negative control. (B) Measurement of binding activity of fluoKDs by ELISA. Symbols: j, KD only; d, N-fluoKD; N, C-fluoKD; ., BSA.

Competitive FLISA and competitive ELISA To validate the application of KD-based FLISA toward an immunoassay tool, the competitive FLISA experiments were carried out and compared with a conventional ELISA system. In both competitive ELISA and FLISA, N-fluoKD or C-fluoKD was incubated with serially diluted target molecule (DR5) on a DR5-coated well plate. After incubation and washing, the remaining fluoKD bound to the DR5 coated was analyzed by direct detection of fluorescent intensity (FLISA) or by HRP-conjugated anti-GFP antibody (ELISA). In the case of competitive FLISA using N-fluoKD, the detectable range of DR5 concentrations was from 100 pg/ml to 64 ng/ml (Fig. 5A). The competitive FLISA using C-fluoKD showed a slightly higher detection range (from 500 pg/ml to 320 ng/ml) than that using N-fluoKD, but it also appeared to be quite sensitive. Competitive ELISA was also performed to compare its sensitivity with that of competitive FLISA. As shown in Fig. 5B, both fluoKDs in competitive ELISA showed much higher detection ranges (from 1.6 to 200 lg/ml)

A

Concentration of DR5 (µ µg/ml)

B

Relative Signal Intensity (A/A 0)

intensity was positively correlated with the concentration of the NfluoKD in the range of 1.28–800 lg/ml (Fig. 4A). In the case of CfluoKD, the good correlation between concentration of C-fluoKD and fluorescent intensity was observed in a similar range of concentration, although it exhibited less fluorescent intensity than N-fluoKD (Fig. 4B). The relatively lower intensities of C-fluoKD might be related not to the binding activity of KD but rather to the relatively lower fluorescent intensity of GFP in the C-fluoKD format (Fig. 3A). In both cases, the use of fluoKD exhibited a fully distinguishable signal against specific antigen DR5 compared with a negative control using BSA. Considering these results, we concluded that the fusion of KD and GFP did not cause any deleterious effect on the fluorescent activity of GFP or on the specific binding activity of KD independent of fusion positions.

67

Relative Signal Intensity (F/F0)

Human kringle domain-based FLISA system / G.M. Jeong et al. / Anal. Biochem. 451 (2014) 63–68

Concentration of DR5 (µg/ml) Fig.5. (A) Competitive FLISA data of N-fluoKD and C-fluoKD. (B) Competitive ELISA data of N-fluoKD and C-fluoKD. Symbols: j, N-fluoKD; ., C-fluoKD.

compared with those in competitive FLISA. Based on competitive FLISA and ELISA, the limits of detection (LODs) for DR5 determination in FLISA using N-fluoKD and C-fluoKD were estimated as 10,000- and 1000-fold lower than those in competitive ELISAs, respectively. These results indicate that the FLISA using fluoKD allows rapid and highly sensitive detection compared with a traditional ELISA system.

Fluorescence intensity (arbitrary unit)

A

Discussion

Concentration of N-fluoKD (µg/ml)

Fluorescence intensity (arbitrary unit)

B

Concentration of C-fluoKD (µg/ml) Fig.4. (A) Conventional FLISA of N-fluoKD (j). (B) Conventional FLISA of C-fluoKD (j). In both experiments, BSA (.) was used as negative control.

Since the last decade, there has been considerable progress in the development of non-immunoglobulin protein scaffolds [12,13]. In addition to having similar properties as antibodies, protein scaffolds have several useful features such as small size (

A human kringle domain-based fluorescence-linked immunosorbent assay system.

As a non-immunoglobulin protein scaffold, human kringle domain (KD) has attractive properties such as high specificity, stability, and production in b...
895KB Sizes 3 Downloads 3 Views